Skip to contentSkip to conversation

sound/soc/codecs: add LAPIS Semiconductor ML26124

PATCH Superseded Tomoya MORINAGA 4 participants 18 messages Nov 28, 2011 – Dec 2, 2011

Subsystems (2)

v3 of 8 Superseded Latest is v8 →

lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 sound/soc/codecs: add LAPIS Semiconductor ML26124

ML26124-01HB/ML26124-02GD is 16bit monaural audio CODEC which has high resistance to voltage noise. On chip regulator realizes power supply rejection ratio (PSRR) be over 90dB so more than 50dB is improved than ever. ML26124-01HB/ ML26124-02GD can deliver stable audio performance without being affected by noise from the power supply circuit and peripheral components. The chip also includes a composite video signal output, which can be applied to various portable device
requirements. The ML26124 is realized these functions into very small package
the size is only 2.56mm x 2.46mm therefore can be construct high quality sound system easily.
ML26124-01HB is 25pin WCSP package; ML26124-02GD is 32pin WQFN package.

---
 sound/soc/codecs/Kconfig   |    4 +
 sound/soc/codecs/Makefile  |    2 +
 sound/soc/codecs/ml26124.c |  475 ++++++++++++++++++++++++++++++++++++++++++++
 sound/soc/codecs/ml26124.h |  122 +++++++++++
 4 files changed, 603 insertions(+), 0 deletions(-)
 create mode 100644 sound/soc/codecs/ml26124.c
 create mode 100644 sound/soc/codecs/ml26124.h
sound/soc/codecs/Kconfig +4 -0
index 665d924..ce8e622 100644
@@ -38,6 +38,7 @@ config SND_SOC_ALL_CODECS
 	select SND_SOC_MAX98095 if I2C
 	select SND_SOC_MAX9850 if I2C
 	select SND_SOC_MAX9877 if I2C
+	select SND_SOC_ML26124 if I2C
 	select SND_SOC_PCM3008
 	select SND_SOC_SGTL5000 if I2C
 	select SND_SOC_SN95031 if INTEL_SCU_IPC
@@ -211,6 +212,9 @@ config SND_SOC_MAX98095
 config SND_SOC_MAX9850
 	tristate
 
+config SND_SOC_ML26124
+	tristate
+
 config SND_SOC_PCM3008
        tristate
 
sound/soc/codecs/Makefile +2 -0
index 5119a7e..c1353ff 100644
@@ -24,6 +24,7 @@ snd-soc-l3-objs := l3.o
 snd-soc-max98088-objs := max98088.o
 snd-soc-max98095-objs := max98095.o
 snd-soc-max9850-objs := max9850.o
+snd-soc-ml26124-objs := ml26124.o
 snd-soc-pcm3008-objs := pcm3008.o
 snd-soc-sgtl5000-objs := sgtl5000.o
 snd-soc-alc5623-objs := alc5623.o
@@ -122,6 +123,7 @@ obj-$(CONFIG_SND_SOC_JZ4740_CODEC)	+= snd-soc-jz4740-codec.o
 obj-$(CONFIG_SND_SOC_MAX98088)	+= snd-soc-max98088.o
 obj-$(CONFIG_SND_SOC_MAX98095)	+= snd-soc-max98095.o
 obj-$(CONFIG_SND_SOC_MAX9850)	+= snd-soc-max9850.o
+obj-$(CONFIG_SND_SOC_ML26124)   += snd-soc-ml26124.o
 obj-$(CONFIG_SND_SOC_PCM3008)	+= snd-soc-pcm3008.o
 obj-$(CONFIG_SND_SOC_SGTL5000)  += snd-soc-sgtl5000.o
 obj-$(CONFIG_SND_SOC_SN95031)	+=snd-soc-sn95031.o
sound/soc/codecs/ml26124.c new file +475 -0

475 lines not on this page · raw message

sound/soc/codecs/ml26124.h new file +122 -0

122 lines not on this page · raw message

Signature · 1 line
1.7.4.4
#1 lore raw
Quoted 93 lines · Tomoya MORINAGA

ML26124-01HB/ML26124-02GD is 16bit monaural audio CODEC which has high resistance to voltage noise. On chip regulator realizes power supply rejection ratio (PSRR) be over 90dB so more than 50dB is improved than ever. ML26124-01HB/ ML26124-02GD can deliver stable audio performance without being affected by noise from the power supply circuit and peripheral components. The chip also includes a composite video signal output, which can be applied to various portable device
requirements. The ML26124 is realized these functions into very small package
the size is only 2.56mm x 2.46mm therefore can be construct high quality sound system easily.
ML26124-01HB is 25pin WCSP package; ML26124-02GD is 32pin WQFN package.

---
 sound/soc/codecs/Kconfig   |    4 +
 sound/soc/codecs/Makefile  |    2 +
 sound/soc/codecs/ml26124.c |  475 ++++++++++++++++++++++++++++++++++++++++++++
 sound/soc/codecs/ml26124.h |  122 +++++++++++
 4 files changed, 603 insertions(+), 0 deletions(-)
 create mode 100644 sound/soc/codecs/ml26124.c
 create mode 100644 sound/soc/codecs/ml26124.h
sound/soc/codecs/Kconfig +4 -0
index 665d924..ce8e622 100644
@@ -38,6 +38,7 @@ config SND_SOC_ALL_CODECS
 	select SND_SOC_MAX98095 if I2C
 	select SND_SOC_MAX9850 if I2C
 	select SND_SOC_MAX9877 if I2C
+	select SND_SOC_ML26124 if I2C
 	select SND_SOC_PCM3008
 	select SND_SOC_SGTL5000 if I2C
 	select SND_SOC_SN95031 if INTEL_SCU_IPC
@@ -211,6 +212,9 @@ config SND_SOC_MAX98095
 config SND_SOC_MAX9850
 	tristate
 
+config SND_SOC_ML26124
+	tristate
+
 config SND_SOC_PCM3008
        tristate
 
sound/soc/codecs/Makefile +2 -0
index 5119a7e..c1353ff 100644
@@ -24,6 +24,7 @@ snd-soc-l3-objs := l3.o
 snd-soc-max98088-objs := max98088.o
 snd-soc-max98095-objs := max98095.o
 snd-soc-max9850-objs := max9850.o
+snd-soc-ml26124-objs := ml26124.o
 snd-soc-pcm3008-objs := pcm3008.o
 snd-soc-sgtl5000-objs := sgtl5000.o
 snd-soc-alc5623-objs := alc5623.o
@@ -122,6 +123,7 @@ obj-$(CONFIG_SND_SOC_JZ4740_CODEC)	+= snd-soc-jz4740-codec.o
 obj-$(CONFIG_SND_SOC_MAX98088)	+= snd-soc-max98088.o
 obj-$(CONFIG_SND_SOC_MAX98095)	+= snd-soc-max98095.o
 obj-$(CONFIG_SND_SOC_MAX9850)	+= snd-soc-max9850.o
+obj-$(CONFIG_SND_SOC_ML26124)   += snd-soc-ml26124.o
 obj-$(CONFIG_SND_SOC_PCM3008)	+= snd-soc-pcm3008.o
 obj-$(CONFIG_SND_SOC_SGTL5000)  += snd-soc-sgtl5000.o
 obj-$(CONFIG_SND_SOC_SN95031)	+=snd-soc-sn95031.o
sound/soc/codecs/ml26124.c new file +24 -0
new file mode 100644
index 0000000..ee323fb
@@ -0,0 +1,475 @@
+/*
+ * Copyright (C) 2011 LAPIS Semiconductor Co., Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307, USA.
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/pm.h>
+#include <linux/i2c.h>
+#include <linux/slab.h>

sound/soc/codecs/ml26124.c line 34

Quoted 10 lines
sound/soc/codecs/ml26124.c new file +10 -0
+#include <linux/platform_device.h>
+#include <sound/core.h>
+#include <sound/pcm.h>
+#include <sound/pcm_params.h>
+#include <sound/soc.h>
+#include <sound/tlv.h>
+#include "ml26124.h"
+
+struct ml26124_priv {
+	struct snd_pcm_substream *substream;

Is substream actually ever used?

Quoted 2 lines
sound/soc/codecs/ml26124.c new file +2 -0
+};
+

[...]

Quoted 52 lines
sound/soc/codecs/ml26124.c new file +52 -0
+
+static const struct snd_kcontrol_new ml26124_output_mixer_controls[] = {
+	SOC_DAPM_SINGLE("DAC Switch", ML26124_SPK_AMP_OUT, 1, 1, 0),
+	SOC_DAPM_SINGLE("Line in Switch", ML26124_SPK_AMP_OUT, 3, 1, 0),
+	SOC_DAPM_SINGLE("PGA Switch", ML26124_SPK_AMP_OUT, 5, 1, 0),
+};
+
+/* Input mux */
+static const char * const ml26124_input_select[] = {"Analog MIC in",
+						    "Digital MIC in"};
+
+static const struct soc_enum ml26124_insel_enum =
+	SOC_ENUM_SINGLE(ML26124_MIC_IF_CTL, 0, 1, ml26124_input_select);
+
+static const struct snd_kcontrol_new ml26124_input_mux_controls =
+	SOC_DAPM_ENUM("Input Select", ml26124_insel_enum);
+
+static const struct snd_kcontrol_new ml26124_line_control =
+	SOC_DAPM_SINGLE("Switch", ML26124_PW_LOUT_PW_MNG, 1, 1, 0);
+
+static const struct snd_soc_dapm_widget ml26124_dapm_widgets[] = {
+	SND_SOC_DAPM_SUPPLY("MCLK", ML26124_CLK_EN, 0, 0, NULL, 0),
+	SND_SOC_DAPM_SUPPLY("PLL", ML26124_CLK_EN, 1, 0, NULL, 0),
+	SND_SOC_DAPM_SUPPLY("MICBIAS", ML26124_PW_REF_PW_MNG, 0, 0, NULL, 0),
+	SND_SOC_DAPM_MIXER("Output Mixer", ML26124_PW_SPAMP_PW_MNG, 0, 0,
+			   &ml26124_output_mixer_controls[0],
+			   ARRAY_SIZE(ml26124_output_mixer_controls)),
+	SND_SOC_DAPM_DAC("DAC", "Playback", ML26124_PW_DAC_PW_MNG, 1, 0),
+	SND_SOC_DAPM_ADC("ADC", "Capture", ML26124_PW_IN_PW_MNG, 1, 0),
+	SND_SOC_DAPM_PGA("PGA", ML26124_PW_IN_PW_MNG, 3, 0, NULL, 0),
+	SND_SOC_DAPM_MUX("Input Mux", SND_SOC_NOPM, 0, 0,
+			  &ml26124_input_mux_controls),
+	SND_SOC_DAPM_SWITCH("Line Out Enable", SND_SOC_NOPM, 0, 0,
+			     &ml26124_line_control),
+	SND_SOC_DAPM_INPUT("VIDEOIN"),
+	SND_SOC_DAPM_INPUT("MDIN"),
+	SND_SOC_DAPM_INPUT("MIN"),
+	SND_SOC_DAPM_INPUT("LIN"),
+	SND_SOC_DAPM_OUTPUT("VIDEOOUT"),
+	SND_SOC_DAPM_OUTPUT("SPOUT"),
+	SND_SOC_DAPM_OUTPUT("LOUT"),
+};
+
+static const struct snd_soc_dapm_route ml26124_intercon[] = {
+	/* output mixer */
+	{"Output Mixer", "PGA Switch", "PGA"},
+	{"Output Mixer", "DAC Switch", "DAC"},
+	{"Output Mixer", NULL, "Line in Switch"},
+	/* outputs */
+	{"SPOUT", NULL, "Output Mixer"},
+	{"LOUT", NULL, "Output Mixer"},
+	{"SPOUT", NULL, "Output Mixer"},

sound/soc/codecs/ml26124.c line 157

Quoted 10 lines
sound/soc/codecs/ml26124.c new file +10 -0
+
+	/* input mux */
+	{"Input Mux", "Analog MIC in", "MICBIAS"},
+	{"Input Mux", "Digital MIC in", "MICBIAS"},
+	{"ADC", NULL, "Input Mux"},
+
+	/* inputs */
+	{"DAC", NULL, "MDIN"},
+	{"PGA", NULL, "MIN"},
+	{"SPOUT", NULL, "LIN"},

The routing looks a bit incomplete. There are quite a few dapm widgets which are not used. E.g. MCLK, PLL, so they will never be enabled, which I would assume would prevent the device from working. Also none of the inputs is routed to the ADC, which seems a bit strange.

Quoted 2 lines
sound/soc/codecs/ml26124.c new file +2 -0
+};
+

[...]

sound/soc/codecs/ml26124.c line 206

Quoted 2 lines
sound/soc/codecs/ml26124.c new file +2 -0
+static int snd_card_ml7213i2s_hw_free(struct snd_pcm_substream *substream,
+				      struct snd_soc_dai *dai)

The name of the function doesn't really match the drivers name.

sound/soc/codecs/ml26124.c line 214

Quoted 8 lines
sound/soc/codecs/ml26124.c new file +8 -0
+{
+	struct snd_soc_codec *codec = dai->codec;
+
+	/* soft reset assert */
+	snd_soc_update_bits(codec, ML26124_SW_RST, 0x01, 1);
+
+	/* Stop  Record/Playback Running */
+	snd_soc_update_bits(codec, ML26124_REC_PLYBAK_RUN, 0x3, 0);

I would expect that this to cause problems with concurrent capture/playback and one of them being stopped.

Quoted 4 lines
sound/soc/codecs/ml26124.c new file +4 -0
+
+	return 0;
+}
+

[...]

Quoted 9 lines
sound/soc/codecs/ml26124.c new file +9 -0
+
+static int ml26124_probe(struct snd_soc_codec *codec)
+{
+	int ret;
+
+	ret = snd_soc_codec_set_cache_io(codec, 8, 8, SND_SOC_I2C);
+	if (ret < 0) {
+		dev_err(codec->dev, "Failed to set cache I/O: %d\n", ret);
+		return ret;

sound/soc/codecs/ml26124.c line 391

Quoted 10 lines
sound/soc/codecs/ml26124.c new file +10 -0
+	}
+
+	/* Software Reset */
+	snd_soc_update_bits(codec, ML26124_SW_RST, 0x01, 1);
+	snd_soc_update_bits(codec, ML26124_SW_RST, 0x01, 0);
+
+	ml26124_set_bias_level(codec, SND_SOC_BIAS_STANDBY);
+
+	snd_soc_add_controls(codec, ml26124_snd_controls,
+			     ARRAY_SIZE(ml26124_snd_controls));

Controls should use table based setup as well.

sound/soc/codecs/ml26124.c line 397

Quoted 6 lines
sound/soc/codecs/ml26124.c new file +6 -0
+
+	return 0;
+}
+
+/* power down chip */
+#define REG_CACHE_SIZE		0x79

You've defined ML26134_CACHESIZE with the same value above.

Quoted 14 lines
sound/soc/codecs/ml26124.c new file +14 -0
+static struct snd_soc_codec_driver soc_codec_dev_ml26124 = {
+	.probe =	ml26124_probe,
+	.suspend =	ml26124_suspend,
+	.resume =	ml26124_resume,
+	.set_bias_level = ml26124_set_bias_level,
+	.reg_cache_size = REG_CACHE_SIZE,
+	.reg_word_size = sizeof(u8),
+	.reg_cache_default = ml26124_reg,
+	.dapm_widgets = ml26124_dapm_widgets,
+	.num_dapm_widgets = ARRAY_SIZE(ml26124_dapm_widgets),
+	.dapm_routes = ml26124_intercon,
+	.num_dapm_routes = ARRAY_SIZE(ml26124_intercon),
+};
+

[...]

#2 lore raw

2011/11/28 Lars-Peter Clausen <[email protected]>:

Quoted 1 line

Also none of the inputs is> routed to the ADC, which seems a bit strange.

This same as my referencing "wm8731.c".Could you give me more information?

Thanks,
Tomoya
ROHM Co., Ltd.

#3 lore raw
Quoted 5 lines · Tomoya MORINAGA

2011/11/28 Lars-Peter Clausen <[email protected]>:

Quoted 1 line

Also none of the inputs is> routed to the ADC, which seems a bit strange.

This same as my referencing "wm8731.c".Could you give me more information?

The DAPM routes are used to do power management based on active path and for this the routes should more or less represent the routing of internal audio signals in your CODEC. I would expect the analog signals to go to the ADC on the CODEC. And if you don't connect them in your DAPM routings ASoC will probably not power-up the ADC since there are no active source signals.

Also this doesn't make much sense:
+ {"Input Mux", "Analog MIC in", "MICBIAS"},
+ {"Input Mux", "Digital MIC in", "MICBIAS"},

You have mux, but both possible settings of the mux connect to the same source, which pretty much makes the mux useless.

#4 lore raw
Quoted 6 lines · Lars-Peter Clausen

Also this doesn't make much sense:

sound/soc/codecs/ml26124.c new file +2 -0
+	{"Input Mux", "Analog MIC in", "MICBIAS"},
+	{"Input Mux", "Digital MIC in", "MICBIAS"},

You have mux, but both possible settings of the mux connect to the same source, which pretty much makes the mux useless.

Plus the fact that the MICBIAS is unlikely to be directly connected to either input within the CODEC, it will almost certainly be connected by circuitry on the board and therefore mapped by the machine driver for the system.

#5 lore raw
In reply to Lars-Peter Clausen

2011/11/29 Lars-Peter Clausen <[email protected]>:

Quoted 8 lines

Also this doesn't make much sense:
+       {"Input Mux", "Analog MIC in", "MICBIAS"},
+       {"Input Mux", "Digital MIC in", "MICBIAS"},

You have mux, but both possible settings of the mux connect to the same source, which pretty much makes the mux useless.

Yes, you are right.
This setting was miss-settings.
This MICBIAS supplies to external MIC as reference voltage. So, I think MICBIAS shouldn't appear at this setting. I show modified settings.

       /* input */
       {"Input Mux", "Analog MIC SingleEnded in", "MIN"},
       {"Input Mux", "Analog MIC Differential in", "MIN"},
       {"Input Mux", "Digital MIC in", "MDIN"},
       {"ADC", "PGA", "Input Mux"},

Thanks,
------
ROHM Co., Ltd.
Tomoya MORINAGA

#6 lore raw
Quoted 15 lines · Tomoya MORINAGA

2011/11/29 Lars-Peter Clausen <[email protected]>:

Quoted 8 lines

Also this doesn't make much sense:

sound/soc/codecs/ml26124.c new file +2 -0
+       {"Input Mux", "Analog MIC in", "MICBIAS"},
+       {"Input Mux", "Digital MIC in", "MICBIAS"},

You have mux, but both possible settings of the mux connect to the same source, which pretty much makes the mux useless.

Yes, you are right.
This setting was miss-settings.
This MICBIAS supplies to external MIC as reference voltage. So, I think MICBIAS shouldn't appear at this setting. I show modified settings.

Quoted 6 lines
       /* input */
       {"Input Mux", "Analog MIC SingleEnded in", "MIN"},
       {"Input Mux", "Analog MIC Differential in", "MIN"},
       {"Input Mux", "Digital MIC in", "MDIN"},
       {"ADC", "PGA", "Input Mux"},

The middle part(the control) of the path must either be a control attached to the mixer on the left side(the sink) or a element of the mux's enum if a mux is on the left side. If the left side is neither a mixer or a mux the middle part is ignored.

The other problem probably is that you have one mux which lets you switch between single-ended analog input, differential analog input and digital input. The first two options are routed through PGA and ADC the later not.

- Lars

#7 lore raw

2011/12/2 Lars-Peter Clausen <[email protected]>:

Quoted 8 lines

The middle part(the control) of the path must either be a control attached to the mixer on the left side(the sink) or a element of the mux's enum if a mux is on the left side. If the left side is neither a mixer or a mux the middle part is ignored.

The other problem probably is that you have one mux which lets you switch between single-ended analog input, differential analog input and digital input. The first two options are routed through PGA and ADC the later not.

I can understand.
The "digital input" description shouldn't be at snd_soc_dpram_route description. So, I've modified like below. These satisfies your showed condition.

	/* Supply */
	{"DAC", NULL, "MCLK"},
	{"ADC", NULL, "MCLK"},
	{"DAC", NULL, "PLL"},
	{"ADC", NULL, "PLL"},

	/* output mixer */
	{"Output Mixer", "PGA Switch", "PGA"},
	{"Output Mixer", "DAC Switch", "DAC"},
	{"Output Mixer", "Line in Switch", "LIN"},

	/* outputs */
	{"LOUT", NULL, "Output Mixer"},
	{"SPOUT", NULL, "Output Mixer"},

	/* input */
	{"Input Mux", "Analog MIC SingleEnded in", "MIN"},
	{"Input Mux", "Analog MIC Differential in", "MIN"},
	{"ADC", NULL, "Input Mux"},

Thanks,
tomoya

#8 lore raw
In reply to Tomoya MORINAGA
Quoted 3 lines · Tomoya MORINAGA

ML26124-01HB/ML26124-02GD is 16bit monaural audio CODEC which has high resistance to voltage noise. On chip regulator realizes power supply rejection ratio (PSRR) be over 90dB so more than 50dB is improved than ever. ML26124-01HB/

To repeat what I said last time please fix the word wrapping in your commit message.

sound/soc/codecs/ml26124.c line 35

Quoted 3 lines
sound/soc/codecs/ml26124.c new file +3 -0
+struct ml26124_priv {
+	struct snd_pcm_substream *substream;
+};

I can't see this being referenced anywhere in the code.

sound/soc/codecs/ml26124.c line 39

Quoted 3 lines
sound/soc/codecs/ml26124.c new file +3 -0
+/* ML26124 configuration */
+static const DECLARE_TLV_DB_SCALE(digital_vol, -7150, 50, 0);
+static const DECLARE_TLV_DB_SCALE(eq_band_gain, -7150, 50, 0);

These are identical.

sound/soc/codecs/ml26124.c line 85

Quoted 4 lines
sound/soc/codecs/ml26124.c new file +3 -0
+	SOC_SINGLE_TLV("EQ Band1 Input Volume", ML26124_EQ_GAIN_BRAND1, 0,
+			0xff, 1, eq_band_gain),
 
+	SOC_SINGLE("EQ BAND1 Switch", ML26124_FILTER_EN, 3, 1, 0),

These should mathc.

sound/soc/codecs/ml26124.c line 90

Quoted 2 lines
sound/soc/codecs/ml26124.c new file +2 -0
+	SOC_SINGLE("Play Limitter Switch", ML26124_DVOL_CTL, 0, 1, 0),
+	SOC_SINGLE("Capture Limitter Switch", ML26124_DVOL_CTL, 1, 1, 0),

Limiter.

sound/soc/codecs/ml26124.c line 92

Quoted 1 line
sound/soc/codecs/ml26124.c new file +1 -0
+	SOC_SINGLE("Digital Volume Switch", ML26124_DVOL_CTL, 4, 1, 0),

Volume Switch makes no sense.

sound/soc/codecs/ml26124.c line 99

Quoted 1 line
sound/soc/codecs/ml26124.c new file +1 -0
+	SOC_DAPM_SINGLE("Line in Switch", ML26124_SPK_AMP_OUT, 3, 1, 0),

Line In.

sound/soc/codecs/ml26124.c line 174

Quoted 8 lines
sound/soc/codecs/ml26124.c new file +8 -0
+       case 16000:
+               snd_soc_update_bits(codec, ML26124_SMPLING_RATE, 0xf, 3);
+               snd_soc_update_bits(codec, ML26124_PLLNL, 0xff, 0x0c);
+               snd_soc_update_bits(codec, ML26124_PLLNH, 0x1, 0);
+               snd_soc_update_bits(codec, ML26124_PLLML, 0xff, 0x20);
+               snd_soc_update_bits(codec, ML26124_PLLMH, 0x3f, 0x00);
+               snd_soc_update_bits(codec, ML26124_PLLDIV, 0x1f, 0x04);
+               break;

You're configuring the PLL here... I assume there's some fixed clock rate that your driver supports, you should at least say what it is.

sound/soc/codecs/ml26124.c line 211

Quoted 7 lines
sound/soc/codecs/ml26124.c new file +7 -0
+static int snd_card_ml7213i2s_hw_free(struct snd_pcm_substream *substream,
+				      struct snd_soc_dai *dai)
+{
+	struct snd_soc_codec *codec = dai->codec;
+
+	/* soft reset assert */
+	snd_soc_update_bits(codec, ML26124_SW_RST, 0x01, 1);

If you need to do this for some reason I'd expect it to be in the bias configuration. Apart from anything else this will run for both playback and capture streams. I'm also not seeing anything that restores the register cache later on.

sound/soc/codecs/ml26124.c line 214

Quoted 2 lines
sound/soc/codecs/ml26124.c new file +2 -0
+	/* Stop  Record/Playback Running */
+	snd_soc_update_bits(codec, ML26124_REC_PLYBAK_RUN, 0x3, 0);

This seems a bit odd immediately after the device was reset...

sound/soc/codecs/ml26124.c line 283

Quoted 2 lines
sound/soc/codecs/ml26124.c new file +2 -0
+#define ML26134_CACHESIZE 79
+static const u16 ml26124_reg[ML26134_CACHESIZE] = {

As previously mentioned you should use regmap.

Quoted 6 lines
sound/soc/codecs/ml26124.c new file +6 -0
+static int ml26124_set_bias_level(struct snd_soc_codec *codec,
+		enum snd_soc_bias_level level)
+{
+	switch (level) {
+	case SND_SOC_BIAS_ON:
+		/* VMID ON */

sound/soc/codecs/ml26124.c line 318

Quoted 10 lines
sound/soc/codecs/ml26124.c new file +10 -0
+		snd_soc_update_bits(codec, ML26124_PW_REF_PW_MNG,
+				    ML26124_VMID, ML26124_VMID);
+	case SND_SOC_BIAS_PREPARE:
+	case SND_SOC_BIAS_STANDBY:
+		break;
+	case SND_SOC_BIAS_OFF:
+		/* VMID OFF */
+		snd_soc_update_bits(codec, ML26124_PW_REF_PW_MNG,
+				    ML26124_VMID, 0);
+		break;

I'd be very surprised if this works well - you're powering up VMID after DAPM has finished and there's no delay at all for bringing up VMID.

sound/soc/codecs/ml26124.c line 327

Quoted 4 lines
sound/soc/codecs/ml26124.c new file +4 -0
+#define ML26124_RATES (SNDRV_PCM_RATE_8000 | SNDRV_PCM_RATE_11025 |\
+		       SNDRV_PCM_RATE_16000 | SNDRV_PCM_RATE_22050 |\
+		       SNDRV_PCM_RATE_32000 | SNDRV_PCM_RATE_44100 |\
+		       SNDRV_PCM_RATE_48000)

This doesn't match the set of rates for which you're configuring magic values at all.

sound/soc/codecs/ml26124.c line 420

Quoted 3 lines
sound/soc/codecs/ml26124.c new file +3 -0
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (priv == NULL)
+		return -ENOMEM;

devm_kzalloc().

#9 lore raw

2011/11/28 Mark Brown <[email protected]>:>> + SOC_SINGLE("EQ BAND1 Switch", ML26124_FILTER_EN, 3, 1, 0),

Quoted 1 line

These should mathc.

I couldn't understand above.
Let me know this meaning.

Thanks

Tomoya
ROHM Co., Ltd.

#10 lore raw
Quoted 5 lines · Tomoya MORINAGA

2011/11/28 Mark Brown <[email protected]>:>> +

Quoted 1 line

These should mathc.

I couldn't understand above.
Let me know this meaning.

Match.

#11 lore raw

2011/11/29 Mark Brown <[email protected]>:> > These should mathc.

Quoted 1 line

Match.

Sorry, still I couldn't understand.
What does your "match" mean for anything?

Thanks

------
ROHM Co., Ltd.
tomoya

#12 lore raw

At Thu, 1 Dec 2011 16:34:14 +0900,

Quoted 7 lines · Tomoya MORINAGA

2011/11/29 Mark Brown <[email protected]>:> > These should mathc.

Quoted 1 line

Match.

Sorry, still I couldn't understand.
What does your "match" mean for anything?

You need to have read Mark's comment carefully. He quoted another line before that.

================================================================

sound/soc/codecs/ml26124.c line 85

Quoted 4 lines
sound/soc/codecs/ml26124.c new file +3 -0
+	SOC_SINGLE_TLV("EQ Band1 Input Volume", ML26124_EQ_GAIN_BRAND1, 0,
+			0xff, 1, eq_band_gain),
 
+	SOC_SINGLE("EQ BAND1 Switch", ML26124_FILTER_EN, 3, 1, 0),

These should mathc.
================================================================

So, you have two inconsistent names for EQ band1.

Takashi

#13 lore raw

2011/12/1 Takashi Iwai <[email protected]>:

Quoted 7 lines

You need to have read Mark's comment carefully.  He quoted another line before that.

================================================================

Quoted 2 lines

+     SOC_SINGLE_TLV("EQ Band1 Input Volume", ML26124_EQ_GAIN_BRAND1, 0,
+                     0xff, 1, eq_band_gain),

Quoted 7 lines
Quoted 1 line

+     SOC_SINGLE("EQ BAND1 Switch", ML26124_FILTER_EN, 3, 1, 0),

These should mathc.
================================================================

So, you have two inconsistent names for EQ band1.

You mean "EQ Band1" and "EQ BAND1" are inconsistent names ? If yes, can you accept the followings ?

EQ Band1 Input Volume
EQ Band1 Switch

Thanks,

------
tomoya
ROHM Co., Ltd.

#14 lore raw
Quoted 5 lines · Tomoya MORINAGA

You mean "EQ Band1" and "EQ BAND1" are inconsistent names ? If yes, can you accept the followings ?

EQ Band1 Input Volume
EQ Band1 Switch

"EQ Band1 Input" and "EQ Band1" are still different, they should be the same.

#15 lore raw

2011/12/1 Mark Brown <[email protected]>:

Quoted 11 lines · Tomoya MORINAGA
Quoted 6 lines

You mean "EQ Band1" and "EQ BAND1" are inconsistent names ? If yes, can you accept the followings ?

EQ Band1 Input Volume
EQ Band1 Switch

"EQ Band1 Input" and "EQ Band1" are still different, they should be the same.

Which description do you want ?

EQ Band1 Volume
EQ Band1 Switch

or

EQ Band1 Input Volume
EQ Band1 Input Switch

or

another

Thanks,
------
tomoya
ROHM Co., Ltd.

#16 lore raw
Quoted 9 lines · Tomoya MORINAGA

Which description do you want ?

EQ Band1 Volume
EQ Band1 Switch

or

EQ Band1 Input Volume
EQ Band1 Input Switch

Either is fine, if I had to pick one I'd go with the first one. The important thing is that Switch and Volume use the same string.

#17 lore raw

2011/12/2 Mark Brown <[email protected]>:

Quoted 7 lines · Tomoya MORINAGA
Quoted 5 lines

Which description do you want ?

EQ Band1 Volume
EQ Band1 Switch

Quoted 8 lines
Quoted 4 lines

or

EQ Band1 Input Volume
EQ Band1 Input Switch

Either is fine, if I had to pick one I'd go with the first one.  The important thing is that Switch and Volume use the same string.

OK, I would use the first one.

Thanks,

tomoya